PDA

View Full Version : Open letter to conference delegates



Mike Whittle
09-12-1994, 01:06 AM
Dear Biomch-l readers:

Last year I posted an "open letter to conference organizers", which was well
received, and has been reprinted in the American Society of Biomechanics
newsletter. I would like to follow it up by offering an "open letter to
conference delegates", which is aimed at improving the quality of oral and
poster presentations at meetings. I realize there are books and seminars on
this topic - what follows is simply my ideas, gathered from quite a few years
of listening to many good presentations - and just as many bad ones!

This may be reproduced and distributed, with suitable acknowledgement.

Mike Whittle
================================================== ===========================
Open Letter to Conference Delegates

by Michael W. Whittle, M.D., Ph.D.
Cline Chair of Rehabilitation Technology
The University of Tennessee at Chattanooga

Avoiding common errors when presenting research

1. Avoid using long complicated words to impress people - it
doesn't! "Eschew obfuscation!!"
2. Avoid spurious accuracy
> Percentages: if there are less than 100 in the sample,
don't give decimal places: 7 out of 11 is 64%, not
63.64%!
> Statistical values such as "p" and "r" should only be
quoted to 2 places, e.g. p give subject ages as mean (to one decimal place) and
range, not standard deviation, for example:
> mean 43.1 years, range 29-68 (easy to understand)
> mean 43.148 years, s.d. 7.415 (hard to understand)
3. When drawing charts, don't let a computer design crazy scales
for you, e.g. 2.19, 5.38, 7.57...
4. Don't assume that everyone reading or listening to your paper
is an expert in the field - make it clear enough for students,
and for people from other disciplines.
5. Understand that there may still be a real difference between
two groups, even if the statistical test fails to support it
(e.g. because the sample was too small).
6. Also understand that a statistically significant result may
have no practical significance in the "real world".
7. Whether it is a written paper, an oral presentation or a
poster, have someone else review it before you go public!

Oral presentations

Planning and preparation
1. Don't simply read a paper that would be suitable for
publication - an oral presentation is a totally different
medium which requires a totally different approach.
2. One picture is worth a thousand words.
3. Humor is useful within limits - it can make a talk more
interesting, but mustn't distract the audience from what you
are trying to say. Sometimes a relevant cartoon will help get
a point across, but a long irrelevant joke will detract from
the talk.
4. Avoid irrelevant slides - nature scenes, glamour pictures,
etc. They may "wake the audience up", but they may also
distract or offend.
5. Unless you are very experienced, do a dry run for timing. It
is better to make it too short than too long - leave them
begging for more, not begging you to stop!
6. Look through the slides after loading the magazine, to make
sure none are backwards or upside down.

Speaking - general

1. Make a point of studying the technique of other speakers when
you go to meetings - learn from both their good and their bad
points.
2. Preferably, talk "off the cuff", using the slides as notes.
If you can't do this, use notes on file cards. If you are too
frightened and must use a script, write it as a speech, not as
a written paper.
3. If you are using notes or a script, make sure there is enough
light to read. If not, try and get a reading light of some
sort.
4. Find out how to use the pointer and control the slides before
you go up to give the talk.
5. Look at each slide as it goes up on the screen - don't just
plough ahead, oblivious of projection problems.
6. Talk to the audience, not to the screen or your notes.
7. Make sure you can be heard - allow for the deaf person in the
back row! If using a microphone, stay about 12 inches from it
and talk normally - don't either stray away from it, or talk
too closely into it.
8. Point to relevant items on the slides. This is difficult if
you are reading a script or heavily dependent on notes.
9. What is on the screen must relate to what you are talking
about - if you want to talk about something different, you
need another slide.
10. Conversely, don't put things on the slide that you don't
intend to talk about - make a simpler slide. "Busy" slides
are a disaster!
11. Stick to the time limit. No matter how interesting it may be,
you will lose your audience's attention if you over-run
significantly. My talks average one slide per minute, so for
a ten minute talk I aim to use ten slides (maybe eleven or
twelve), but certainly not fifteen, twenty, or more!
12. Tell your audience clearly when you have finished. Don't say
"Well, that's about it..." or "Any questions?" Thank them for
their attention and wait for the applause!

Overhead projection

1. Always look at the screen to make sure the image is
satisfactory.
2. Don't block the projector beam.
3. Don't block the audience's view. An overhead projector tends
to obstruct the view anyway, and you standing beside it makes
things worse!
4. Don't "fiddle" with the overhead transparency or with
something you are using as a pointer or to cover up part of
the text.
5. Only cover up part of the overhead if there is a good reason
to do so - otherwise, it is somewhat insulting to the
audience!
6. The safest technique is to use overhead transparencies like
slides - stand by the screen and point to the projected image.
7. Normal typescript is not big enough to show up when projected:
use an enlarging photocopier or presentation graphics.

Slide projection

1. Don't use dual slide projection just for the sake of it.
2. Advantages of dual projection:
> it allows you to mix illustrations and caption slides
> it allows you to show a lecture plan as you go through
the topics
3. Disadvantages of dual projection:
> it doubles (or more) the chance of something going wrong
> it makes the presentation twice as complicated (and
costly) to prepare
> it gives you a problem when you only want to show
something on one projector
4. If using dual projection, keep the projectors synchronized.
5. If possible, avoid black backgrounds (stick-type pointers
don't show up) or white backgrounds (light pointers don't show
up).

Poster presentations

1. Don't put too much on the poster - it deters people from
stopping to read it.
2. Have a large clear title, name the authors, and
institution(s).
3. Don't write a poster all in capitals (except the title) - it
makes it hard to read. Use 1 1/2 or 2 line spacing, and keep the
sentences and paragraphs short.
3. Have an abstract or summary which says briefly what you did
and what you found.
4. Don't write it like a paper - use the same "visual" approach
as for a talk, with charts rather than tables (but no jokes!).
5. Don't assume everyone will read it from start to finish - they
won't! Every part of it must be complete and self-
explanatory.
6. Don't "pounce" on anyone who comes near the poster, but be
ready to talk people through it if they are interested.
7. Don't be disappointed if nobody comes to look at your poster -
it happens to everybody!